CAPÍTULO I: PLANTEAMIENTO DEL PROBLEMA
1.4 Alcances y limitaciones
DEFINITION OF OBJECTIVES AND SCOPE OF THE PSA PROJECT
3.1. Paragraphs 3.1 and 3.2 provide recommendations on meeting Requirement 4 of Ref. [3] on purpose of the Level 1 PSA and Requirement 14 relating to the scope of a Level 1 PSA [3]. Determination of the objectives of the PSA together with its intended and potential uses is an important step to undertake prior to starting the process of performing a PSA. The scope of the PSA is defined by the analysis level (Level 1, 2 or 3), the initiating events and hazards considered, and the operational modes (i.e. full power, low power or shutdown states6) addressed. The scope of the PSA should be compatible with both the objectives of the study and the available resources and information, i.e. the necessary procedures and methods, personnel, expertise, funding and the time needed for the analysis. For example, if the objective of a PSA is to verify the risk arising from plant operation against specified safety goals, thus implying a complete risk assessment, a full scope PSA comprising a comprehensive listing of initiating events and hazards and all plant operational modes should be performed, and adequate resources should be provided for the analysis. In addition, other sources of radiation (e.g. the spent fuel pool) may need to be analysed, depending on the formulation of the safety goals. 3.2. It should be recognized that the intended applications of PSA may impose additional requirements on the scope of the PSA, on the modelling approaches and on the level of detail. If such additional requirements are taken into account at the planning stage of the PSA project, it will help to avoid inconsistencies in the results and insights obtained. For instance, if it is planned to use the PSA for the development of a severe accident management programme, a Level 2 PSA should be performed. As another example, if it is planned to use the PSA model as a basis for a risk monitor, the PSA model should be ‘symmetrical’ in terms of the modelling of initiating events. The common simplification of modelling an
initiating event as always occurring in one particular train should not be used. For example, loss of coolant accidents should be modelled for each loop with an appropriate probability that a specific loop is affected (i.e. 1/2 for a two train plant, 1/3 for a three train plant) rather than a single event in one of the loops. More details on the features of PSA necessary for various applications of PSA are provided in Section 10.
PROJECT MANAGEMENT FOR PSA
3.3. Paragraphs 3.3–3.14 provide recommendations on meeting Requirement 5 of Ref. [3] on preparation for the safety assessment for Level 1 PSA and on meeting Requirement 22 of Ref. [3] on management of the safety assessment. Project management of the PSA depends strongly on the specific conditions in a State, namely:
(a) The organizations participating in the PSA project;
(b) The type and extent of the involvement of the participating organizations; (c) The objectives and the scope of the PSA study.
After the objectives and the scope of the PSA have been specified, the management scheme for the PSA project should be developed, including the selection of methods and establishment of procedures, the selection of personnel and the organization of the team that will perform the PSA, the training of the team, the preparation of a PSA project schedule, the estimation and securing of the necessary funds, and the establishment of quality assurance procedures and peer review procedures.
3.4. A PSA study is normally commissioned by one of the following: (a) The plant designer;
(b) The operating organization of the plant; (c) The regulatory body.
The PSA can be performed by these groups or by consultants, research institutes, universities or a combination of these. In any case, the operating organization should always participate as a source of operational knowledge, as well as being a beneficiary from the insights obtained.
3.5. It is generally considered desirable to start the process of performing the PSA as early as possible in the lifetime of the plant. Design weaknesses or
procedural weaknesses that are recognized early can be corrected or improved less expensively than those that remain until the plant is in operation. While a PSA can be started in any of the stages in the lifetime of the plant, the PSA models and documentation should be maintained and regularly updated throughout the operating life of the plant to provide continued benefit.
3.6. The PSA study should consider a particular ‘freeze date’ for modelling the as built and as operated plant conditions. If it is known at the beginning of the PSA project that certain changes in plant design and operation will be implemented in the near term, before the PSA is finished, a decision should be taken at an early stage of the PSA as to whether these changes will be addressed in the PSA. If the decision is made to address the future changes, the freeze date should be determined accordingly and the PSA should take account of the status of the plant after the modifications.
3.7. The documentation for the PSA should be developed in a clear, traceable, systematic and transparent manner so that it can effectively support the review of PSA, applications of PSA and future PSA upgrades.
SELECTION OF METHODS AND ESTABLISHMENT OF PROCEDURES 3.8. Appropriate working methods and procedures should be established at the outset of the project so that there is a minimum of modification to these procedures during the project. Unnecessary iterations in methods and procedures may cause delays in the PSA project. General guidance for the methodological tools and approaches to analysis is given in the following sections of this publication. Once the working methods have been selected, the various procedural steps should be interfaced with the tasks of quality assurance and training to produce a detailed plan of the tasks, including a schedule for the project.
3.9. The resources in terms of the expertise of the specialists involved, human resources, computer time, calendar time and so on that will be necessary to complete a PSA depend greatly on the scope of the PSA, which is in turn governed by the overall objectives, and on the available expertise in the PSA team. Scheduling of the activities should be carried out following the establishment of detailed procedures and should account for the availability of personnel.
TEAM SELECTION AND ORGANIZATION
3.10. The members of the team that perform the PSA can be characterized by the organization they represent and the technical expertise they provide. Once the necessary personnel have been identified, lines of communication should be set up and specific tasks should be assigned. The training necessary should be determined and planned in accordance with the activities of the PSA. The task of team formation and training is closely associated with the corresponding tasks of quality assurance.
3.11. The expertise necessary to conduct a PSA should provide two essential elements: knowledge of the plant and knowledge of PSA techniques. This expertise can vary in depth, depending on the scope of the PSA, but the participation of the plant designer and the operating organization of the plant should be foreseen, if possible. More specifically, the necessary expertise relating to knowledge of the plant should be obtained from persons with extensive familiarity with the design and operation of the plant under normal and accident conditions.
3.12. A team that will perform a PSA for the first time should be provided with training to acquire the expertise necessary to complete the study successfully.
ESTABLISHMENT OF A QUALITY ASSURANCE PROGRAMME FOR PSA 3.13. The quality assurance7 programme for a PSA encompasses activities that are necessary to achieve the appropriate quality of the PSA and activities that are necessary to verify that the appropriate quality is achieved. For a PSA, appropriate quality means an end product that is correct and usable and one which meets the objectives and fulfils the scope of the PSA. The quality assurance programme should provide for a disciplined approach to all activities affecting the quality of the PSA, including, where appropriate, verification that each task has been satisfactorily performed and that necessary corrective actions have been implemented.
7 Instead of the term ‘quality assurance’, the term ‘management system’ is used in Ref. [7]. The term ‘quality assurance’ is left in this Safety Guide in order to comply with widely accepted current practices and terminology used in the area of PSA.
3.14. Quality assurance of the PSA should be viewed and established as an integral part of the PSA project and the quality assurance procedures should be an integral part of the PSA procedures. The quality assurance procedures should provide for control of the constituent activities associated with a PSA in the areas of organization, technical work and documentation. In their application to the technical work, quality assurance procedures are aimed at ensuring consistency between goals, scope, methods and assumptions, as well as accuracy in the application of methods and in calculations. Quality assurance procedures should include control of the documentation of the PSA. General requirements for control of documents are established in section 2 of Ref. [7].
GENERAL ASPECTS OF PSA DOCUMENTATION
Objectives and content of documentation
3.15. Paragraphs 3.15–3.22 provide general recommendations on meeting Requirement 20 on documentation for Level 1 PSA [3]. The primary objectives of the PSA documentation should be to fulfil the requirements of its users and be suitable for the specific applications of the PSA. Possible users of the PSA include:
(a) Operating organizations of nuclear power plants (management and operating personnel);
(b) Designers and vendors;
(c) Regulatory bodies and persons or organizations providing them with technical support;
(d) Other government bodies; (e) The public.
Some of these users, the public for example, might use, primarily, the summary report of the PSA, while others will use the full PSA documentation, including the computer model.
3.16. PSA documentation includes work files, computer inputs and outputs, correspondence, interim reports and the final report of the PSA. The documentation of PSA should be complete, well structured, clear and easy to follow, review and update. It should be presented in a traceable and sequential manner, i.e. the order of appearance of analysis in the final documentation should follow, as far as possible, the order in which it was actually performed. In addition, means should be provided for possible extensions of the analysis,
including integration of new topics, use of improved models, broadening of the scope of the PSA in question and its use for alternative applications. Explicit presentation of the assumptions, exclusions and limitations for extending and interpreting the PSA is also of critical importance to users.
3.17. The documentation should provide within the report (or by reference to available material) all necessary information to reconstruct the results of the study. All intermediate subanalyses, calculations, assumptions, etc., that will not be published in any external reports should be retained as notes, working papers or computer outputs. This is very important for reconstructing and updating each detail of the analysis in the future.
Organization of documentation
3.18. The final report of the PSA study should be divided into three major parts: (1) Summary report;
(2) Main report;
(3) Appendices to the main report.
3.19. The summary report should be designed to provide an overview of the motivations, objectives, scope, assumptions, results and conclusions of the PSA at a level that is useful to a wide audience of reactor safety specialists and that is adequate for high level review. The summary report is designed:
(a) To support high level review of the PSA;
(b) To communicate key aspects of the study to a wide audience of interested parties;
(c) To provide a clear framework and guide for the reader or user prior to consulting the main report.
The summary report of a PSA should include a subsection on the structure of the report, which should present concise descriptions of the contents of the sections of the main report and of the individual appendices. The relation between various parts of the PSA should also be included in this subsection of the summary report. 3.20. The main report should give a clear and traceable presentation of the complete PSA study, including a description of the plant, the objectives of the study, the methods and data used, the initiating events considered, the plant modelling results and the conclusions. The main report, together with its appendices, should be designed:
(a) To support technical review of the PSA;
(b) To communicate key detailed information to interested users;
(c) To permit the efficient and varied application of the PSA models and results;
(d) To facilitate the updating of the models, data and results in order to support the continued safety management of the plant.
3.21. The appendices should contain detailed data, records of engineering computations, detailed models, etc. The appendices should be structured so as to correspond directly to the sections and subsections of the main report, as far as possible.
3.22. In addition to the general recommendations for documentation provided in this section, specific recommendations for documentation are provided in other sections of this Safety Guide, for example, for PSA for internal initiating events, for PSA for internal fire, for PSA for internal flooding, for PSA for external hazards and for PSA for low power and shutdown states.